quantitative high-speed imaging of entire developing ... · quantitative high-speed imaging of...
TRANSCRIPT
1
Quantitative High-Speed Imaging of Entire Developing Embryos
with Simultaneous Multi-View Light Sheet Microscopy
Raju Tomer, Khaled Khairy, Fernando Amat and Philipp J. Keller*
Howard Hughes Medical Institute, Janelia Farm Research Campus *Correspondence: [email protected]
Supplementary Information
Supplementary Figures
Supplementary Figure 1 Technology framework for simultaneous multi-view imaging
Supplementary Figure 2 Optical implementation of the SiMView imaging platform
Supplementary Figure 3 SiMView real-time electronics and computational hardware
Supplementary Figure 4 Point-spread-functions in one- and two-photon SiMView
Supplementary Figure 5 Spatio-temporal artifacts in sequential multi-view imaging
Supplementary Figure 6 Quantifying Drosophila nuclei dynamics in the syncytial blastoderm
Supplementary Figure 7 Simultaneous multi-view imaging of Drosophila embryogenesis
Supplementary Figure 8 Post-acquisition larval hatching in sequential multi-view imaging
Supplementary Figure 9 Manual cell tracking in the retracting germ band
Supplementary Figure 10 Reconstruction of neuroblast and epidermoblast lineages
Supplementary Figure 11 Time-course of C155-GAL4,UAS-mCD8::eGFP signal intensity
Supplementary Figure 12 SiMView optical slices of the Drosophila VNC and brain lobes
Supplementary Tables
Supplementary Table 1 Components of the one-photon SiMView light sheet microscope
Supplementary Table 2 Components of the two-photon SiMView light sheet microscope
Supplementary Table 3 Specifications of simultaneous multi-view imaging experiments
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 1 | Technology framework for simultaneous multi-view imaging
2
Nature Methods: doi:10.1038/nmeth.2062
3
Supplementary Figure 1 | Technology framework for simultaneous multi-view imaging
The SiMView framework for simultaneous multi-view imaging consists of a light sheet
microscopy platform with four independent optical arms, a real-time electronics control
framework and three computational modules for dual-sCMOS high-speed image acquisition,
high-throughput content-based multi-view image fusion and high-throughput image data
management. A fourth post-processing pipeline performs fast and accurate image segmentation
and cell tracking in SiMView live recordings of nuclei-labeled specimens.
This figure provides an overview of the key capabilities and functions of the six core modules,
which are described in more detail in the Online Methods. The opto-mechanical components,
electronics and computational hardware (marked with asterisks) are described in Supplementary
Figs. 2 and 3, Supplementary Tables 1 and 2 and in the Online Methods. Core modules of the
computational framework for high-throughput multi-view image registration and image fusion
are provided as Supplementary Software.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 2 | Optical implementation of the SiMView imaging platform
4
Nature Methods: doi:10.1038/nmeth.2062
5
Supplementary Figure 2 | Optical implementation of the SiMView imaging platform
(a) Computer model of the opto-mechanical implementation of the light sheet microscope for
simultaneous multi-view imaging. The opto-mechanical modules of the instrument consist of two
illumination arms for fluorescence excitation with scanned light sheets (blue), two fluorescence
detection arms equipped with sCMOS cameras (red) as well as beam-coupling modules,
specimen chamber and the specimen positioning system (grey). Each illumination sub-system
comprises the following elements (listed according to appearance along the beam path, from
beam coupling unit to specimen chamber): custom collimator module, mirror pair, illumination
filter wheel, laser shutter, piezo tip/tilt mirror, f-theta lens, tube lens and piezo-mounted
illumination objective. Each detection sub-system comprises the following elements (listed
according to appearance along beam path from specimen chamber to detector): piezo-mounted
detection objective, detection filter wheel, tube lens, tube system, sCMOS camera. Detailed
information on all of the microscope’s hardware components, including optics, mechanical parts,
electronics and computer hardware, are provided in Supplementary Tables 1 and 2.
(b) Enlarged view of the central part of the microscope, where the four optical arms for
simultaneous multi-view imaging meet at the specimen chamber. The specimen positioning
system is located underneath the specimen chamber, which provides three essential advantages:
full access to the specimen chamber from the top, sufficient space for opto-mechanical
components in the four-arm arrangement surrounding the sample chamber, and mechanical
specimen support by an upright specimen holder. The latter feature is required for specimen
embedding in ultra-low concentration agarose gels for physiological long-term imaging
experiments.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 3 | SiMView real-time electronics and computational hardware
6
Nature Methods: doi:10.1038/nmeth.2062
7
Supplementary Figure 3 | SiMView real-time electronics and computational hardware
This diagram shows the real-time electronics framework for precise control and relative timing
of all opto-mechanical components of the SiMView imaging platform, as well as the integrated
computational hardware for high-speed image acquisition and high-throughput multi-view image
registration, image fusion, wavelet compression and image data management. Custom LabVIEW
software communicates the process workflow to the NI real-time controller and coordinates data
streaming on the image acquisition workstation. Custom Matlab software performs high-
throughput image analysis on the processing workstation and manages the data stream to the
storage server. Core modules of the high-throughput content-based multi-view image registration
and image fusion pipeline are provided as Supplementary Software.
Optionally, an additional image processing workstation with a Tesla GPU (Nvidia Corporation)
is integrated in the framework to enable high-throughput image segmentation and cell tracking
using a custom image processing pipeline (see Online Methods).
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 4 | Point-spread-functions in one- and two-photon SiMView
8
Nature Methods: doi:10.1038/nmeth.2062
9
Supplementary Figure 4 | Point-spread-functions in one- and two-photon SiMView
(a) Representative lateral (x,y) and axial (x,z) views of 50-nm-sized fluorescent beads recorded
with the one-photon light sheet microscope for simultaneous multi-view imaging (one-photon
SiMView). The microscope was equipped with 40x Carl Zeiss Plan-Apochromat water-dipping
detection objectives (NA = 1.0), representing the optical configuration in recordings shown in
Supplementary Videos 16-20. Average FWHM values resulted as 0.399 ± 0.025 µm laterally,
and 1.59 ± 0.13 µm axially (mean ± s.d., n = 5). A false color look-up-table was used to enhance
visualization.
(b) Representative lateral (x,y) and axial (x,z) views of 50-nm-sized fluorescent beads recorded
with the two-photon light sheet microscope for simultaneous multi-view imaging (two-photon
SiMView). The microscope was equipped with 16x Nikon water-dipping detection objectives
(NA = 0.8), representing the optical configuration in recordings shown in Supplementary
Videos 4-6. Average FWHM values in the center of the field-of-view (FOV) resulted as 0.603 ±
0.086 µm laterally, and 1.87 ± 0.14 µm axially (mean ± s.d., n = 8). Average FWHM values at the
edge of the field-of-view (FOV) resulted as 0.640 ± 0.026 µm laterally, and 2.29 ± 0.06 µm
axially (mean ± s.d., n = 2). Note that lateral FWHM values are larger than in (a) owing to the
lower sampling (and thus lower effective imaging resolution) obtained with the 16x detection
objective. A false color look-up-table was used to enhance visualization.
Scale bars, 1 µm.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 5 | Spatio-temporal artifacts in sequential multi-view imaging
10
Nature Methods: doi:10.1038/nmeth.2062
11
Supplementary Figure 5 | Spatio-temporal artifacts in sequential multi-view imaging
(a) Maximum-intensity projections of the front and back halves of a fused sequential four-view
light sheet microscopy recording of a stage 4 Drosophila embryo just after the 13th mitotic cycle.
To obtain the sequential multi-view panels, image fusion was performed on a subset of the
simultaneous multi-view data set, corresponding to a 105 seconds delay between the first and last
views.
(b) Corresponding projections of the fused simultaneous four-view light sheet microscopy
recording of the same embryo at the same time point.
(c) Enlarged views of the regions indicated by white rectangles in (a). Shape information and
temporal patterns are distorted in the sequential multi-view recording. Moreover, nuclei
undergoing division are often accompanied by ghost images in the sequential multi-view
recording.
(d) Maximum-intensity projections of the front half of a sequential (left) and a simultaneous
(right) multi-view recording of the same embryo as in (a) during the 13th mitotic cycle. The delay
between the first and last views in the sequential multi-view recording is 105 seconds.
(e) Enlarged views of the region indicated by a white rectangle in (d). In the sequential multi-
view recording, nuclei appear in duplicates (or triplicates, if the nucleus is undergoing a
division), which represent the same nucleus at different locations and in different mitotic states.
Scale bars, 50 µm (a,b,d), 20 µm (c,e).
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 6 | Quantifying Drosophila nuclei dynamics in the syncytial blastoderm
12
Nature Methods: doi:10.1038/nmeth.2062
13
Supplementary Figure 6 | Quantifying Drosophila nuclei dynamics in the syncytial blastoderm
Quantitative analysis of the 12th and 13th mitotic waves performed on the segmentation and
tracking results shown in Supplementary Video 8. All statistics were obtained with Huber
robust estimator (implemented and described in the Matlab function robustfit) in order to avoid
bias from outliers arising from tracking inaccuracies as specified in the methods section
“Quantitative Estimation of Segmentation and Tracking Accuracy”.
(a) Distribution of nuclei speeds directly after nuclear division. Mean and standard deviation are
8.12 ± 2.59 µm min-1 (n = 2,798) for mitotic wave 12 and 7.21 ± 2.21 µm min-1 (n = 4,852) for
mitotic wave 13. These distributions represent the statistics used to obtain the values at t = 0.42
min (global maxima) in Fig. 5d.
(b) Average distance between daughter nuclei from the same mother nucleus as a function of
time t after nuclear division. As in Fig. 5d, the small standard error arises from the large sample
size of ~2,500-5,000 samples per time point. The average distance between daughter nuclei
reaches a maximum of 11.05 µm at t = 1.25 min after division for mitotic wave 12 and 10.07 µm
at t = 1.67 min after division for mitotic wave 13, which is almost two-fold higher than the
global average nearest neighbor distance of post-mitotic nuclei in the embryo shown in Fig. 5e.
Subsequently, 10 minutes after division, the average distance between daughter nuclei relaxes to
8.76 µm for mitotic wave 12 and to 5.68 µm for mitotic wave 13, owing to the almost two-fold
increase in nuclei count by the end of each mitotic wave.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 7 | Simultaneous multi-view imaging of Drosophila embryogenesis
14
Nature Methods: doi:10.1038/nmeth.2062
15
Supplementary Figure 7 | Simultaneous multi-view imaging of Drosophila embryogenesis
Maximum-intensity projections of image stacks from a one-photon SiMView time-lapse
recording of Drosophila embryonic development. Separate background-corrected maximum-
intensity projections of the first and second halves of the fused three-dimensional image stacks
are shown for eight time points, providing dorsolateral and ventrolateral views of the developing
embryo. Owing to the activation of the embryonic muscular system towards the end of the
recording, fast contractions are constantly reorienting the entire embryo (such as in the last panel,
where the embryo is turned by approximately 90 degrees). Simultaneous multi-view imaging
allows following development even through this technically particularly challenging period. The
entire embryo was recorded in 35-second intervals over a period of 19.5 hours, using an image
acquisition period of 15 seconds per time point. The complete recording comprises one million
images (10 terabytes) for ~2,000 time points recorded from 2 to 21.5 hours post fertilization
(Supplementary Video 3). A second recording of a different embryo from 3 to 18.5 hours post
fertilization is provided as well (Supplementary Video 2).
PC = pole cells, VF = ventral furrow, eGB/rGB = extending/retracting germ band, CF = cephalic
furrow, A = amnioserosa, VNC = ventral nerve cord, BL = brain lobes.
Scale bar, 50 µm.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 8 | Post-acquisition larval hatching in simultaneous multi-view imaging
16
Nature Methods: doi:10.1038/nmeth.2062
17
Supplementary Figure 8 | Post-acquisition larval hatching in simultaneous multi-view imaging
(a) Image of the sample cylinder showing the hatched larva after long-term high-speed image
acquisition of a developing embryo with the light sheet microscope. An ellipsoidal-shaped
hollow space is visible in the 0.4% low-melting-temperature (LMT) agarose cylinder protruding
from the glass capillary and indicates the previous location of the embryo during image
acquisition. The larva shown here hatched from the embryo recorded in Supplementary Video
12. A short time sequence of the behaving larva is provided in Supplementary Video 7.
(b) Enlarged view of the hatched larva after time-lapse imaging of the C155-GAL4,UAS-
mCD8::GFP embryo in Supplementary Video 12.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 9 | Manual cell tracking in the retracting germ band
18
Nature Methods: doi:10.1038/nmeth.2062
19
Supplementary Figure 9 | Manual cell tracking in the retracting germ band
Manual tracking was performed for cells in a non-superficial layer of the retracting germ band,
confirming that SiMView recordings also provide the spatio-temporal resolution required for
quantitative analyses of cellular dynamics in late embryonic stages.
4D manual tracking was performed for a period of 113 time points in the two-photon SiMView
recording presented in Supplementary Video 5, using the ImageJ Manual Tracking plug-in
(http://rsbweb.nih.gov/ij/plugins/track/track.html). The spatial tracking coordinates were
converted into a Vaa3D-compatible data format and rendered using Vaa3D version 2.7 (Peng et
al., Nature Biotechnology, 2011). This figure shows an example of a manually reconstructed cell
track with a total length of approximately 250 µm, superimposed with maximum-intensity
projections of the fused SiMView data set. Note that the two-dimensional Vaa3D visualization
highlights the rendered track on top of the maximum-intensity projection and thus gives the
impression of a superficial trajectory. The three-dimensional cell location, however, is in a non-
superficial layer of the germ band throughout the analyzed time window (optical depth between
32.5 µm at t0 and 44.7 µm at t112).
(a) Maximum-intensity projection of the two-photon SiMView recording at time point t6,
superimposed with the rendered track between t0 and t6.
(b) Maximum-intensity projection of the two-photon SiMView recording at time point t56,
superimposed with the rendered track between t0 and t56.
(c) Maximum-intensity projection of the two-photon SiMView recording at time point t112, superimposed with the rendered track between t0 and t112.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 10 | Reconstruction of neuroblast and epidermoblast lineages
20
Nature Methods: doi:10.1038/nmeth.2062
21
Supplementary Figure 10 | Reconstruction of neuroblast and epidermoblast lineages
(a) Raw optical slices from SiMView recording in Supplementary Video 3 for key events in the
lineage reconstructions visualized in Fig. 6a,b and Supplementary Video 11. Optical slices
indicate blastoderm origins, delamination, first cell division and second cell division for three
neuroblasts, as well as blastoderm origin and first cell division for one epidermoblast. Yellow
arrows indicate the locations of the nuclei of the tracked cells. The appearance of stripes in the
raw data arises from the column gain variability typically encountered in first generation sCMOS
cameras (such as the Andor Neo detector used in this recording). The SiMView processing
pipeline contains a module for measuring column gain factors and correcting these stripes.
(b) Lineage trees for the neuroblast/epidermoblast lineage reconstructions visualized in Fig. 6a,b
and Supplementary Video 11 (1st div. = first division, 2nd div. = second division). Four
blastoderm cells and their respective daughter cells were manually tracked from time point 0 to
400 (120-353 minutes post fertilization, 35 second temporal resolution), using Imaris (Bitplane)
and ImageJ (http://rsbweb.nih.gov/ij/). Tracks start in the blastoderm (time point 0). The
neuroblasts delaminate between time points 227 and 251, and subsequently produce ganglion
mother cells in two division cycles (first cycle between time points 310 and 332, second cycle
between time points 368 and 390). The epidermoblast remains in the outer cell layer and divides
once at time point 313. Manual tracking was performed until time point 400 for all cells.
Scale bar, 10 µm.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 11 | Time-course of C155-GAL4,UAS-mCD8::eGFP signal intensity
22
Nature Methods: doi:10.1038/nmeth.2062
23
Supplementary Figure 11 | Time-course of C155-GAL4,UAS-mCD8::eGFP signal intensity
The figure shows the time course of average GFP signal intensity in a high-speed live recording
of the developing nervous system (Supplementary Video 13). Simultaneous multi-view
imaging started shortly after the onset of GFP expression at around 9.5 hours post fertilization
(time point 0). Approximately 400,000 images of the developing nervous system were recorded
over the period 9.5-15 hours post fertilization. Despite continuous high-speed image acquisition,
GFP signal intensities were constantly increasing. The fluctuations towards the end of the plot
represent signal changes in the percent-range, which arise from re-orientation of the embryo after
the onset of global muscle contractions.
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Figure 12 | SiMView optical slices of the Drosophila VNC and brain lobes
24
Nature Methods: doi:10.1038/nmeth.2062
25
Supplementary Figure 12 | SiMView optical slices of the Drosophila VNC and brain lobes
(a) Maximum-intensity projection of an image stack from the simultaneous multi-view time-
lapse recording of the Drosophila embryonic nervous system shown in Supplementary Video
12 (one-photon SiMView recording at 14.5 hours post fertilization).
(b) Optical slices from the image stack visualized in (a) show the part of the ventral nerve cord
(VNC) highlighted by the white rectangle. The images demonstrate that one-photon SiMView
resolves the cell bodies with an average diameter of 2-3 µm and thus achieves cellular resolution
for the majority of the VNC. Images were deconvolved with the Lucy-Richardson algorithm (5
iterations).
(c) Maximum-intensity projection of an image stack from the simultaneous multi-view time-
lapse recording of the Drosophila embryonic nervous system shown in Supplementary Video
13 (one-photon SiMView imaging at 14.5 hours post fertilization).
(d) Sub-stack maximum-intensity projection of the data set visualized in (c), showing one of the
brain lobes (BL). The sub-stack covers 1/3 of the z-range of the complete stack and comprises
the entire brain lobe.
(e) Optical slices from the image stack visualized in (c) and (d) show the brain lobe highlighted
by the white rectangle in (d). The images demonstrate that one-photon SiMView resolves cell
bodies and thus achieves cellular resolution for a large fraction of the brain lobe. Images were
deconvolved with the Lucy-Richardson algorithm (5 iterations).
Scale bars, 25 µm.
Nature Methods: doi:10.1038/nmeth.2062
26
Supplementary Table 1 | Components of the one-photon SiMView light sheet microscope
Module Component Product(s) Manufacturer
Lasers
(shared modules)
SOLE-6 module Solid-state lasers: 488/642/685
nm
DPSS lasers: 515/561/594 nm
Omicron Laserage
SOLE-3 module Solid-state lasers: 405/445 nm Omicron Laserage
Illumination sub-systems
(two mirrored modules)
High-speed laser shutter VS14S2ZM1-100
with AlMgF2 coating
VMM-D3 three-channel driver Uniblitz
Illumination filter wheel
96A351 filter wheel
MAC6000 DC servo controller Ludl
NDQ neutral density filters Melles Griot
Miniature piezo tip/tilt mirror
S-334 tip/tip mirror
E-503.00S amplifier
E-509.S3 servo controller
E-500 chassis
Physik Instrumente
F-theta lens S4LFT4375 Sill Optics
Tube lens module U-TLU-1-2 camera tube Olympus
Piezo objective positionerP-725 piezo
E-665 piezo amplifier and servo controller
Physik Instrumente
Illumination objective XLFLUOR 4x/340/0.28 Olympus
Detection sub-systems
(two mirrored modules)
Piezo objective positionerP-725 piezo
E-665 piezo amplifier and servo controller
Physik Instrumente
Detection objective
CFI60/75 LWD water-dipping series
Nikon
Apochromat/Plan-Apochromat water-dipping series
Carl Zeiss
Detection filter wheel
96A354 filter wheel
MAC6000 DC servo controller Ludl
RazorEdge and EdgeBasic long-pass filters
BrightLine band-pass filters Semrock
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 1 (continued)
27
Module Component Product(s) Manufacturer
Detection sub-systems
(two mirrored modules)
Tube lens module
CFI second lens unit Nikon
AxioImager 130 mm ISD tube lens
Carl Zeiss
Camera
Neo sCMOS
ExosII re-circulator Andor/Koolance
OrcaFlash Hamamatsu
Specimen chamber with perfusion
system
Four-view specimen chamber
Scaffold manufactured from black Delrin
Custom design
Specimen holder
Holder manufactured from medical-grade stainless steel
Multi-stage adapter module for connection to specimen
positioning system
Custom design
Perfusion system
REGLO dual-channel 12-roller pump
Harvard Apparatus
Model 107 benchtop environmental chamber
TestEquity
Specimen positioning system
Customized translation stages
(three units) M-111K046
Physik Instrumente
Rotary stage M-116 Physik
Instrumente
Motion I/O interface and amplifier
C-809.40 4-channel servo-amplifier
Physik Instrumente
Motion controller PXI-7354 4-axis stepper/servo
motion controller National
Instruments
Real-time electronics
Real-time controller with LabVIEW Real-
Time OS PXI-8110 Core 2 Quad 2.2 GHz
National Instruments
I/O interface boards
(three units) PXI-6733 high-speed analog
output 8-channel board National
Instruments
BNC connector boxes
(three units) BNC-2110 shielded connector
block National
Instruments
Serial interface board PXI-8432/2 National
Instruments
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 1 (continued)
28
Module Component Product(s) Manufacturer
Control software
Real-time modules 32-bit LabVIEW code Custom software
Host modules 64-bit LabVIEW code
C-compiled Matlab code Custom software
Workstations and servers
Data acquisition workstation
2x X5680 HexaCore CPUs Intel Corporation
18x 8 GB DDR-3 RAM modules Kingston
24-channel RAID controller 52445 Adaptec
24x Cheetah 15K.7 SAS-2 600GB hard disks
Seagate
10 Gigabit fiber network adapter EXPX9501AFXSR
Intel Corporation
GeForce GTX470 graphics card Nvidia
Corporation
2x Neon CameraLink frame grabbers BitFlow
2x PCIe-1429 Full-Configuration CameraLink frame grabbers
National Instruments
X8DAH+-F server board Supermicro
Fiber network-attached storage server
S5520SC-based standard rack-mount server
Intel Corporation
Fiber network-attached RAID systems
2x ESDS A24S-G2130 24-disk enclosure
Infortrend
48x Ultrastar A7K2000 SATA-2 2TB hard disks
Hitachi
Nature Methods: doi:10.1038/nmeth.2062
29
Supplementary Table 2 | Components of the two-photon SiMView light sheet microscope
Module Component Product(s) Manufacturer
Lasers
(shared modules)
Pulsed IR laser Chameleon Ultra II Coherent
SOLE-3 module Solid-state lasers: 488/561/594 nm Omicron Laserage
Illumination sub-systems
(two mirrored modules)
Beam attenuation module
AHWP05M-980 mounted achromatic half-wave plate (690-1200 nm)
GL10-B Glan-laser polarizer (650-1050 nm)
Thorlabs
Pockels cell Model 350-80-LA-02 KD*P series
electro-optic modulator
Model 302RM driver Conoptics
IR beam splitter
PBSH-450-2000-100 broadband polarizing cube beam splitter
Melles Griot
WPA1312-2-700-1000 achromatic 1/2 wave plate
Casix
High-speed laser shutter
VS14S2ZM1-100 with AlMgF2 coating
VMM-D3 three-channel driver Uniblitz
Illumination filter wheel
96A351 filter wheel
MAC6000 DC servo controller Ludl
NDQ neutral density filters Melles Griot
Miniature piezo tip/tilt mirror
S-334 tip/tip mirror
E-503.00S amplifier
E-509.S3 servo controller
E-500 chassis
Physik Instrumente
F-theta lens 66-S80-30T-488-1100nm custom lens Special Optics
Tube lens module U-TLU-1-2 camera tube Olympus
Piezo objective positioner
P-622.1CD PIHera piezo linear stage
E-665 piezo amplifier and servo controller
Physik Instrumente
Illumination objective
LMPLN5XIR 5x/0.10
LMPLN10XIR 10x/0.30 Olympus
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 2 (continued)
30
Module Component Product(s) Manufacturer
Detection sub-systems
(two mirrored modules)
Piezo objective positioner
P-622.1CD PIHera piezo linear stage
E-665 piezo amplifier and servo controller
Physik Instrumente
Detection objective
CFI60/75 LWD water-dipping series Nikon
Apochromat/Plan-Apochromat water-dipping series
Carl Zeiss
Detection filter wheel
96A354 filter wheel
MAC6000 DC servo controller Ludl
BrightLine short-pass and band-pass filters
Semrock
Tube lens module CFI second lens unit Nikon
AxioImager 130 mm ISD tube lens Carl Zeiss
Camera Neo sCMOS
ExosII re-circulator Andor/Koolance
Specimen chamber with perfusion
system
Four-view specimen chamber
Scaffold manufactured from black Delrin
Custom design
Specimen holder
Holder manufactured from medical-grade stainless steel
Multi-stage adapter module for connection to specimen positioning
system
Custom design
Perfusion system
REGLO dual-channel 12-roller pump
Harvard Apparatus
Model 107 benchtop environmental chamber
TestEquity
Specimen positioning system
Customized translation stages
(three units) M-111K046
Physik Instrumente
Rotary stage M-116 Physik
Instrumente
Motion I/O interface and
amplifier C-809.40 4-channel servo-amplifier
Physik Instrumente
Motion controller PXI-7354 4-axis stepper/servo
motion controller National
Instruments
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 2 (continued)
31
Module Component Product(s) Manufacturer
Real-time electronics
Real-time controller with LabVIEW Real-Time OS
PXI-8110 Core 2 Quad 2.2 GHz National
Instruments
I/O interface boards
(three units) PXI-6733 high-speed analog
output 8-channel board National
Instruments
BNC connector boxes
(three units) BNC-2110 shielded connector
block National
Instruments
Serial interface board PXI-8432/2 National
Instruments
Control software
Real-time modules 32-bit LabVIEW code Custom software
Host modules 64-bit LabVIEW code
C-compiled Matlab code Custom software
Workstations and servers
Data acquisition workstation
2x X5680 HexaCore CPUs Intel
Corporation
18x 8 GB DDR-3 RAM modules Kingston
24-channel RAID controller 52445 Adaptec
24x Cheetah 15K.7 SAS-2 600GB hard disks
Seagate
10 Gigabit fiber network adapter EXPX9501AFXSR
Intel Corporation
GeForce GTX470 graphics card Nvidia
Corporation
2x Neon CameraLink frame grabbers
BitFlow
2x PCIe-1429 Full-Configuration CameraLink frame grabbers
National Instruments
X8DAH+-F server board Supermicro
Fiber network-attached storage server
S5520SC-based standard rack-mount server
Intel Corporation
Fiber network-attached RAID systems
2x ESDS A24S-G2130 24-disk enclosure
Infortrend
48x Ultrastar A7K2000 SATA-2 2TB hard disks
Hitachi
Nature Methods: doi:10.1038/nmeth.2062
32
Supplementary Table 3 | Specifications of simultaneous multi-view imaging experiments
Experiment
Drosophila
syncytial blastoderm
(specimen #1)
Drosophila
syncytial blastoderm
(specimen #2)
Related figures Figs. 2a-e, 3b,c; 5c
Suppl. Fig. 5
Figs. 2f, 5a,b,d,e
Suppl. Fig. 6
Related videos Video 10 Videos 1, 8, 9
Imaging framework One-photon SiMView
Excitation wavelength 488 nm
Illumination objectives Olympus XLFLUOR 4x/340/0.28
Detection filters Semrock RazorEdge LP488 U-grade
Detection objectives Carl Zeiss C-Apochromat
10x/0.45 W Nikon CFI75 LWD
16x/0.8 W
Cameras Hamamatsu Orca Flash 2.8 Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period 1.94 hours
201 time points
1.61 hours
233 time points
Temporal sampling 35 seconds 25 seconds
Recording period per time point
15 seconds 10 seconds
Exposure time 20 milliseconds 25 milliseconds
Voxel size 0.36 x 0.36 x 2.00 µm3 0.41 x 0.41 x 2.03 µm3
Images per time point 4 x 125 4 x 120
Total # of images 100,500 111,840
Size of data set 518 gigabytes 1.12 terabytes
Post-acquisition fusion 5-level Daubechies D4
wavelet fusion 20-pixel linear blending
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 3 (continued)
33
Experiment
Drosophila
embryonic development
(specimen #1)
Drosophila
embryonic development
(specimen #2)
Related figures Fig. 3d Figs. 3a, 6a
Suppl. Figs. 7, 10a
Related videos Video 2 Videos 3, 11
Imaging framework One-photon SiMView
Excitation wavelength 488 nm
Illumination objectives Olympus XLFLUOR 4x/340/0.28
Detection filters Semrock RazorEdge LP488 U-grade
Detection objectives Nikon CFI75 LWD
16x/0.8 W
Cameras Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period 17.08 hours
2051 time points
19.44 hours
2001 time points
Temporal sampling 30 seconds 35 seconds
Recording period per time point
15 seconds
Exposure time 25 milliseconds
Voxel size 0.41 x 0.41 x 2.03 µm3
Images per time point 4 x 130 4 x 125
Total # of images 1,066,520 1,000,500
Size of data set 10.73 terabytes 10.06 terabytes
Post-acquisition fusion 20-pixel linear blending 20-pixel linear blending
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 3 (continued)
34
Experiment
Drosophila
embryonic development
(specimen #3)
Drosophila
embryonic development
(specimen #4)
Related figures Fig. 4b
Suppl. Fig. 9 Fig. 4c
Related videos Video 5 Video 6
Imaging framework Two-photon SiMView
Excitation wavelength 940 nm
Illumination objectives Olympus LMPLN10XIR 10x/0.3
Detection filters Semrock BrightLine SP680
Semrock BrightLine BP525/50
Detection objectives Nikon CFI75 LWD
16x/0.8 W
Cameras Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period 1.42 hours
171 time points
2.71 hours
326 time points
Temporal sampling 30 seconds
Recording period per time point
20 seconds
Exposure time 150 milliseconds
Voxel size 0.41 x 0.41 x 2.03 µm3
Images per time point 2 x 110 2 x 105
Total # of images 37,620 68,460
Size of data set 387 gigabytes 705 gigabytes
Post-acquisition fusion 4-pixel linear blending
(axially only)
4-pixel linear blending
(axially only)
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 3 (continued)
35
Experiment Drosophila
stage 16 slicing series
Related figures Fig. 4a
Related videos Video 4
Imaging framework Two-photon SiMView
Excitation wavelength 940 nm
Illumination objectives Olympus LMPLN10XIR 10x/0.3
Detection filters Semrock BrightLine SP680
Semrock BrightLine BP525/50
Detection objectives Nikon CFI75 LWD
16x/0.8 W
Cameras Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period Single time point
Temporal sampling −
Recording period per time point
−
Exposure time 200 milliseconds
Voxel size 0.41 x 0.41 x 0.41 µm3
Images per time point 2 x 593
Total # of images 1,186
Size of data set 12 gigabytes
Post-acquisition fusion 3-pixel linear blending
(axially only)
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 3 (continued)
36
Experiment
Drosophila
neural development
C155-GAL4
(specimen #1)
Drosophila
neural development
C155-GAL4
(specimen #2)
Related figures Fig. 6c,d
Suppl. Figs. 8, 12a,b Suppl. Figs. 11, 12c-e
Related videos Videos 7, 12, 15 Videos 13, 14
Imaging framework One-photon SiMView
Excitation wavelength 488 nm
Illumination objectives Olympus XLFLUOR 4x/340/0.28
Detection filters Semrock RazorEdge LP488 U-grade
Detection objectives Nikon CFI75 LWD
16x/0.8 W
Cameras Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period 5.83 hours
701 time points
5.47 hours
788 time points
Temporal sampling 30 seconds 25 seconds
Recording period per time point
15 seconds
Exposure time 25 milliseconds
Voxel size 0.41 x 0.41 x 2.03 µm3
Images per time point 4 x 140 4 x 130
Total # of images 392,560 409,760
Size of data set 3.95 terabytes 4.12 terabytes
Post-acquisition fusion 20-pixel linear blending 20-pixel linear blending
Nature Methods: doi:10.1038/nmeth.2062
Supplementary Table 3 (continued)
37
Experiment
Drosophila
neural development
Ftz-ng-GAL4
(specimen #1)
Drosophila
neural development
Ftz-ng-GAL4
(specimen #2)
Drosophila
neural development
Ftz-ng-GAL4
(specimen #3)
Related figures Fig. 6e − −
Related videos Videos 16, 17, 19 Video 18 Video 20
Imaging framework One-photon SiMView
Excitation wavelength 488 nm
Illumination objectives Olympus XLFLUOR 4x/340/0.28
Detection filters Semrock RazorEdge LP488 U-grade
Detection objectives Carl Zeiss Plan-Apochromat
40x/1.0 W
Cameras Andor Neo sCMOS
Temperature 25.0°C
Sample embedding 0.4% low-melting temperature agarose in tap water
Imaging period 1.37 hours
165 time points
8.33 hours
1,001 time points
2.13 hours
256 time points
Temporal sampling 30 seconds
Recording period per time point
15 seconds
Exposure time 25 milliseconds
Voxel size 0.16 x 0.16 x 2.03 µm3
Images per time point 4 x 110 4 x 115 4 x 105
Total # of images 72,600 460,460 107,520
Size of data set 748 gigabytes 4.63 terabytes 1.08 terabytes
Post-acquisition fusion 20-pixel linear
blending 20-pixel linear
blending 20-pixel linear
blending
Nature Methods: doi:10.1038/nmeth.2062